Skip to main content

Full text of "Thesis [microform]"

See other formats




Collection de 





C.n.diin InstituM for Historical MIcroraproduetiont / Institut Canadian da microraproductiona historiqu 

Technical and Bibliographic Notes / Notes technique et bibliographiques 

The Institute has attempted to obtain the best original 
copy available for filming. Features of this copy which 
may be bibliographically unique, which may alter any of 
the images in the reproduction, or which may 
significantly change the usual method of filming are 
checked below. 

Coloured covers / 
Couverture de couleur 

I I Covers damaged / 

' — ' Couverture endommagee 

I I Covers restored and/or laminated / 
— ' Couverture restauree et/ou pelliculee 

I I Cover titte missing / Le titre de couverture manque 

I I Coloured maps / Cartes g^raptiiques en couleur 

r~| Coloured ink (i.e. other than blue or black) / 

Encre de couleur (i.e. autre que bleue ou noire) 

I I Coloured plates and/or illustrations / 
— ' Planches et/ou Illustrations en couleur 

I I Bound with other material / 

Relte avec d'autres documents 




Only edition available / 
Seule edition disponlble 

Tight binding may cause shadows or distortion 
along interior margin / La reliure serree peut 
causer de I'ombre ou de la distorsion le long de 
la marge int^rieure. 

Blank leaves added during restoratkms may appear 
within the text. Whenever possible, these have 
been omitted from filming / II se peut que certaines 
pages blanches ajouttes tors d'une restauration 
apparaissent dans le texte, mais, kxsque cela dtait 
possible, ces pages n'ont pas M Nmdes. 

L'Instltut a microfilm* le meilleur examplaire qu'il lui a 
ete possible de se procurer. Les details de cet exem- 
plaire qui sont peut-6tre uniques du point de vue bibli- 
ographlque, qui peuvent modifier une image reproduite, 
ou qui peuvent exiger une modifications dans la meth- 
ode normale de filmage sont indiqu^s ci-dessous. 

I I Cokjured pages/ Pages de couleur 

I I Pages damaged / Pages endommagees 

I ] Pages restored and/or laminated / 
— Pages restaurtes et/ou pelteultes 

r^ Pages discoloured, stained or foxed / 
'^^ Pages decolorees, tachetees ou piquees 

I I Pages detached / Pages d^tachees 

r^ Showthrough / Transparence 

I [ Quality of print varies / 

' — ' Quality inegale de I'impression 

I I Includes supplementary material / 

Comprend du materiel suppl^mentaire 

I I Pages wholly or partially obscured by errata 
slips, tissues, etc., have been refllmed to 
ensure the best possible image / Les pages 
totalement ou partlellement obscurcies par un 
feulllet d'enata, une pelure, etc.. ont 6t6 filmtes 
a nouveau de fapon a obtenir la mellleure 
image possible. 

I I Opposing pages with varying colouration or 
discolourations are filmed twice to ensure the 
best possible image / Les pages s'opposant 
ayant des colorations variables ou des decol- 
orations sont film^es deux fois afin d'obtenir la 
meilleur image possible. 

AddKtonaJ comments / 
Commentaires supplementaires: 

Pagination is as follans : p. [10], [«M]-610. 

This ittm is filmad at th* raduetion ratio etiacked lialMr/ 









rMuction indi^iii ci-dtnous 









28 X 


Tha copy filmad hart hai baan raproducad thanki 
to iha ganaroiily of: 

National Library of Canada 

L'axamplaira film* fut raproduit gric* k la 
Qtnttotixt da: 

Blbllotheque natlonale du Canada 

Tha imagat appaaring hara ara tha batt quality 
poatlbia eonaidarlng tha condition and lagibillty 
of tha original copy and in Icaaping with tha 
filming contract (pacificationa. 

Lai imagat auivantat ont txt raproduitat avac la 
plus grand (oin. compta tanu da la condition at 
da la nanatt da l'axamplaira film*, at an 
conformity avac laa conditiona du contrat da 

Original eopiaa in printad papar eovara ara fllmad 
baginning with tha front eovar and anding on 
tha laat paga with a printad or illuatratad impraa- 
sion. or tha back eovar whan appropriata. All 
othar original eopiaa ara filmad baginning on tha 
first paga with a printad or illuatratad Impraa- 
aion, and anding on tha laat paga with a printad 
or illuatratad impraation. 

Laa aiamplairaa originaui dont la eouvartura an 
papiar aat Imprimta sort tilma* an commancant 
par la pramiar plat at an tarminant loit par la 
darnlAra paga qui eompona una amprainta 
d'Imprasaion ou d'illuatration. aoit par la tacond 
plat, aalon la caa. Toua laa autraa axamplairat 
originaui lont fllmta an commandant par la 
pramMra paga qui comporta una amprainta 
d'impraasion ou d'illuatration at an tarminant par 
la darnitra paga qui eompona una talla 

Tha laat racordad frama on aach microficha 
ihall contain tha lymbol — » (moaning "CON- 
TINUED"), or tha aymbol y (moaning "END"), 
whichavar appliaa. 

Un daa aymboloa auivanta apparaitra sur la 
darniira imaga da ehaqua microficha, talon la 
cat: la tymbola ^^ tignifia "A SUIVRE", la 
aymbola ▼ tignifia "FIN". 

Mapa, plataa, eharta, ate. may ba filmad at 
diffarant raduction ratios. Thota too larga to ba 
antiraly includad in ona axpotura ara filmad 
baginning in tha uppar laft hand cornar, laft to 
right and top to bottom, aa many framaa aa 
raquirad. Tha following diagrama illuatrata tha 

Laa cartaa, planchaa, tablaaux, ate., pauvant itra 
fllmta t daa taux da rtduction diffirants. 
Lorsqua la documant asi trop grand pour atra 
raproduit an un aaul clicha, 11 att film* * partir 
da I'angla tuptriaur gaueha. da gaucha * droita. 
at da haut an baa. an pranant la nombra 
d'imagaa nacataaira. Lat diagrammat tuivanta 
illuatrant la mathoda. 










MKaocorr iesoiution tist cha>t 






m^ IB 


S^ ^^^3 Eaal Wain Streal 

S^S Rochsalsr. rtaw York 14609 USA 

•■JS (716) ♦82 - 0300 - Phone 

^B (716) 288-a<)09 -Fai 








Sudmittcd lor th« dtgrec oJ Doctor of Philosophy in tli» 
Univcnity of Toronto 







To the 

We, ihe examiners appointed by the Board of 
Postgraduate Studies for tlie consideration of the theses 
submitted by Mr. J. F. MacKey for the degree of Ph. D.. 
are of the opinion that they are a distinct contribution 
to Chemical knowledge and beg to recommend that they 
be accepted for the degree of Doctor of Philosophj in 
this University. 

W. R. Lang 

T. L. Walker 

W, H. Ellis 

F. B. Allan 

W. Lash Miller 
May 6, 1909. A. P. Colen:an 

I hereby certify that the theses above mentioned 
have been accepted by the Senate of the University of 
Toronto for degree of Doctor of Philosophy in accordance 
with the terms of the statute in that behalf. 



The University of Toronto, 
June 9, 1909 X 





Si'dmitUil ior till ifnt oi Doctor of Philotopky in tkt 
UnivcTUty of Toronto 


1 ^ o ■ 

* j^ 


Bjr WlLLIAU RoBUT IjANH, .IiMlN TRVM tN M.\t KEV. Qllil 


J.M. CrnftM (Hull. Sor. vhim., Is7(i, |iil, U, MU). i" r>-t<Triii;: t>> th.' 
oxiitonce uf odmpmmiU of arNciiii- wiili ' -<ihol nulii-li's, |Htiriti'il nut thnt 
mi i-AtrrM of nriicnioui or ot ursciilc b. ■ ' tuitl U't'ii [>ri'|Miri' \ v\i tn tlir 
yrnr IS7n. nml 'Ifucribnl it nn-tltod Iiy whit-h lu' nbtHinr><l ctlijl, mi-Jhyl. an i 
uniyl arncnati'M; tliis ci'iNiMtpit in h<-iitinu ttio i-urrcNpdnilirt}: intli'!.-^ in 
Bealetl tiibet Ht lOn" with nurnial h'iWvt orwrnti'. The cHriTu vn-t ■ [mrifli-'i 
by watihint; witli othrr »n<i tlintillin): uinitT (litnininhcil pri'^'* ir,'. Thrv 
nfp liquids nf hit^h hoitinK point, above 200 . iirul Huft'ir liiTiiiupoit: 
tiun if iliititjpd uniter utniuiiphrrti- [ncHiiiin-; in tlic m< i>t tin' amy I 
arspnatf ao much ho that ho loiiml it imponiiiblr to obta... it puri' t-vrn wlu-n 
distilled in a vaeuuin. Tin' iulditiDn of Wiitcr to thrm crui' ^< i'mncUal" 
and eomplote decomposition into arHenir acid and the aU'Dhol. Crafts deter- 
mined their comimsitl'in by >,ei;;hini,' the precipitati'tl iiriicriir arid 
at the maKnesiuni salt and the carbon and hydrot;f>n by coinh.i-ition, i > 
gave them the formu'a -R^'AsO^. For the am-nitt's of the iilkvl radicl 
he used three methods, a fcealod tulK* beint; employed: (I) Ihr inti-raction 
of ancniouB oxide and ethyl silicate at 200°, (2) the intrr-rtiou of ethyl 
iodide and silver arsenite at 150°; and (:i) the* interaetiori of arsenic 
bromide and sodium etboxide, which l:ist method he considere ! the best, 
although in it a secondary reaction l>etweeii the ester formed and the 
sodium ethoxide causes n condition of equilibrium to be set up, and no 
more ester is formed. 

It is to be noted that even if disodium hydrogen Aiwuite i» employed 
the resulting arsenite is always the trialkyi salt Crafts states that arsen- 
ious oxide and alcohol do not react when heaicd" f n a sealcii tube, nor d »e»i 
arseniouB oxide either with ether alone or with the addition of ethyl acetate, 
even when in contact for twenty hours. 

These esters prepared by Crafts are specified by Auger (Compt. rend, 
1902, 134, 238) as the only ones known, and he refers in u biter paper 
{Compt. rend,, 1906, 143, 907) to Crafts' inability to obtain any ester 
by heating the alcohols and oxides of arsenic in a sealed tube. Auger 
bases his conviction that an ester is produced on the fact that arsenious 
oxide is volatile in alcohol \-apour, although not in water vapour, and 
concludes that this cannot be accounted for unless there is an ester formed. 
Also he succeeded iu preparing an ester from glycerol and arsenic oxide, 

».!.! lr„n ethyr, ni.thyl. ,V.opr.,pj-|, ,-,„l,utyI, and isoaniyl al.ohols and ar- 
.,n,n„, .,x,do. He L«,tod tho „!.„1,„1, „.ith oostallinc arsenioua o,ido for 
sunu. hours ei.dently in a sealfd tube., a, tho ttM^poratares quoted are far the point, of the ale, hol», a,„l jives 6.5 per eent, a» extreme 
l™,t of e,.ter,ficat,oa in the ease of methyl alcohol, and 1.2 per eent in 
I.:.. of ethyl alcohol. Calculating fron, his figures, the yield of ester 
tor the higher alcohols appears to l,e 2.G2 per cent., 0.23 per eent I 00 
per cent., and ,1.6:) per cent, (or propyl, i.opropyl, i,„!,utyl, and ,*oa.nyl 

So far as can be gathore.l fron, the pap^-r, Auger takes no account of 
.le extreme .so.ulohty, aa we have found, of arsenious oxide in the ester 
. ese small yields and the difficulty found by Crafts in obtaining any re- 

...l.s front the acfon of the acid on the alcohol are evidently due to the 

ic\ers>l ot the reaction: 

(;R()H_|_A»„<J_:iaK.,AsO _|-:ill O . . (1) 

.■istililc' ■■,";■'""»,""" '"' T"^" ""' "■'"" "' " "•'" *■"""'''' ^y ''««"<"■»' 

n xture of alcohol and water vapour over calcium c.rbide placed in an 
"e la,': """' '"";';?« "" "■'"" «■■■' """"■••"g "■<■ -kohol to drop back into 
, n t s'.„ed t ' T""!'"" ""^ '•"'°''- •''" °''^"""'"' ^'"■W. "ow much 
Phro, ,.' ■"■""' "■'""^' ""'' '""'""y' "^^"■"'^ ™'' 'l>us obtained. 

1 h nol, ,n the same wry, yielded an ester hitherto only prepared by the 
act.ou 01 so.liun. phenoxide on arsenic trichloride ^ 

The nature of the reaction expressed by equation (1) and the pro- Of the ester clearly show that only by removing the wat as t 
formed can one expect to Ret a large yield. Our work has been done 
.1 two ways, namely, by heating the mixture in a vessel to which is 
attached a modified Soxhlet tube and condenser, the Soxhlet co^tl ninl 
nhydrons copper sulphate, and by adding this dehydrating atnt d.rect y 
the ntxture. A comparison of the yields obtained i^ given U the 
sequel. By the«i two meth,ds yields of fron, 14.8 per cent (Jo d) to 58 6 
per cent, (hot, have been obtained with aliphatic'alcohols, in^w th th! 
phenols as as 90 per cent. Kxperiments have also b en mal wUh 
^ry encourag,ng results, on ben.y. alcohol and on esters of hydroxy a da 
We hope to show that the use of such a dehydrating agent arconner ! ' 
phate ,„ a Soxhlet tube will allow of the preparation o? a ser Jof s.t 
from the oxides and possibly sulphides of arsenic, antimony tin ,„,! 
perhaps bismuth with compounds containing liydrox;i where the ho', 
pomts of such esters are higher than that of water I im e"d i^h "^ 

wUh all of these; also with anhydrons copper su^L.e'i „ a^ IdthX 
nbstances themselves. In many preparations, too, of esters gen r„| ' an. 
aMied products, the formation of water in uie reaction produces an eonf 
brium resulting in a very low yield| It is expected t'h at by r mol^ 
this water, aa is done in the instances described in the orese,,. '""'"«^ 
yelds will be greatly increased in many eommerei^ll prrs,::.' '""' ''" 

-We have pas«!d the vapour of methyl alcohol over ar,enio,-s „,,.-. 
Wed^in . . tube, and the ether obtained differs entirely Z::^ 


JleaiUg nith Inverted Condcnf:cr C»»/i'.— Wei^licil (luanlUiod of pripyl, 
Mobutyl, and isoamyl alcohols iverc iiiixod with cxct'ss i»f arsciiioiis oxi;lc 
and heated with a direct flame for forty-live minutes in a fluak to v.)iU-h a 
reflux condenser was attached. Tlie clear liquid obt;iincd was iwiiiirl off 
from the excess of arsenious rtxide and fractionated under liiniiiiiahed 
pressure. The esters so purified v.pre aDalys^d and the yield enmp'ir;';) wltli 
that calculated from equation (1). 

Our method of ascertaining tl.e composition of the esters diff'iTs from 
that used by Auger." Tho ester was deoompoaed witli water, forming 
arsenious oxide and liberatinji the alcohol. Ihe urai'nious oxido was 
dissolved in sodium c:irl)oinite and titrated with .V/IO iodine (a preliminary 
lest having shown that the alcohol did not n'act with iodine). By this 
means the quantity of arsenic in the ester was tleternitned. 

i^oAmyl Arsiiiite. — 1.4 Grams were ili-composed with 10 c.c. ot water, 
sufficient sodium carbonate added to dissolve the ai'.^';iiou.s oxide, aud tht- 
whole diluted to 100 c.c; 25 e.e. of th-s r'q.iired 17 c.c of stiindnrd iodine 
for oxidation, corresponding with ^2.4 per ccut. ot arsenic. In 50 ^* 
of the ester, the alcohol was determined aud found ti» wei^h 89 grams, or 
78 per cent. woButyl and propyl arsenltes were also analysed in a similar 
way : 

Per cent. Per cent. As calculated 
Ester As found. ns (KO). As. 

Methyl 44.5 44.6 

Ethyl — — 

Propyl 29.6 l!!>.7 . ' 

isoButyl 25.7 25.5 

Amyl 22,2 22.:'. 

isoAmyl 22.4 22.:; 

Expressing these as salts of arseniims acid, the foriiiK!:)!- Irf-com;* 

Propyl argenitc is a yellow, mobile liquid boiling at 216° and decompos- 
ing very readily on add tion of water. 

iaoButyl nrstcnite is a deep yellow, mobile liqu'd of spi-c-'ic gravity 
1.069; it decomposes rapidly in presence of watrr iiit.i j>o])Utyl alcahol an 1 
arsenioua oxide; under 760 mm. pressure it decomposes at 2-12'', and 'n lils 
at 157" under 30 mm. pressure. 

\soAmyl arnenite is a yellow liquid of specific gravity 1.050: it l-oil* 
at 185° under 30 mm. pressure, and under atmospheric press-.ite ir de- 
composes at 284°; in the presence of water, it dcconip^/i;'^ into /ioamyl 
nlcobol and arsenious oxide. 

*Augpr determined the amount of arsenic in the cooled liq-.iid obtii;n(^d (a 
mixture of the ester and alcohol) by means of iodine. Tho difference be- 
tween this and the amount of arsenious nxido that would di:*:*'.)Ive iu the 
same quanity of alcohol in thp cold was taken as represt-nting the amount 
of ester formed {Compt. rend., 1906, 143, 90S). Xo account was taken of 
the solubility of arsenicnis oxide in the esters themselves. 

for.m.d alis„rbp,l bv flip «„l.,. I, bcjlinsr, and the mUer 

after forty.flve minute,, when ,-„.ll rl, 7 '", "' ""■' "'""^'^ 

'- p. e.t. Ht„:„.. .,t^:^::"„:::r-: z;::,::zr-' -■■ 

three ::l.^Jt"Z:^tZ "' ^"'""•^"'° -'"i«™ti„r o'f tho^e 

if a .'^'o-..rati„; ,,,.:» ■:■'„."", \.:„,r:ru,e"'':, "tt '- '-^ ™"' 

in a stoppered bottle at ro„„, temperature 3, «n '" ' "''"■ ''"""■" 

""d 70 .rams of anhydron, eonner., 1 V. ! """"' "' '"■'""""^ oxide 

yield, were cMainedf: L y.Tr en jl ! 2 "' '""" '"'^»' ^"^ ""'"»-« 
..!« per eent.; propy, ar.e„fte;;L"";„':™t.''" """'■' '""'''''^ "-"^"•• 
ferelr„:;::dr ""■" '"^"^"■' ""■' --P"-"" ^ie,d. (per eent.) b, dif. 

Methy, . ^^"^'i':'-) -'^— . in8„,h,et, iotheeCd ^ 
Ethyl J 2 - 33.8* _ 

^'^ri :::::::::■■ l^ - -- u. 

Trimethylcarbinol ' 1 "'-^S 15.8 

Arnyl ; • J ^ - 54.27 _ 

"oAmyl . , . Qg, 54.00 _ 

condenser. .irsen.ous oxide in a flask ftted with a reflux 

A probable explanation of the low vMH i„ ,i 
lies in the faet that ethyl alcohol iseVh' / ' ""' '" ""■ ""'.vl arsenite 
for small quantities of water p in ' eanlt be""' """'' '"■«'"'" "'■'='■ 
-ulphate. This is really shown^y :« "ra ,roV r'T "' ""■ ''"'■^'""- 
.leoho, in which a little anhydroL copper suphate I, "" "' "' "'■^" 
copper sulphate is apparently n„t .ffee^d wh t )l' " P'"""'^- ^he 
alcohols a bine color appears almost instartirh '" V'"' "' ""' •"«•■" 
copper .„phate. Methyl alcohol ter^t^/^^-^.f ^.-'^ti"" of the 
do the h.,her alcohols, bnt it respond .„ ^V^ TZ^ ^TZ 

Iiighor alcohols, but it respomta more rea.lily tliaii does ethyl alouhol. The 
yield in each ease, using the copper sulphate, probably bears a direct rela- 
Uun to the respective affinities of the alcohol aii.l of the copper sulphate 
for the water of the reaction. If ., substanci' eoul.l be useil which had a 
greater dehydrating power and at the same time was insoluble in alcohol, 
a higher yield would doubtless be obtained. I'alcium carbide (Auger' 
fompl. rend., 1906, 143, 90.S) is not such a material; we l,av,. compared the 
yields obtained by replacing the anhydrous copper sulphate in the Soxlilet 
with fresh calcium carbide, using various alcohols, and the highest yield 
was in the case of i/<oamyl alcohol where 40.2 per cent, of ester was obtained 
after two hours' heating, l l 

Estm obtained uUh I'lu.iol ami ,(» Uoinolooms.^F^T the esteriflcation 
of these with arscnious oxide a side necked llask was use.l the neck of which 
was bent upwards and attache.l to the Soxhlet, leaving the iriouth of the 
dask tree to receive a thermomet;'r. 

Phiiint .Irat-nile.— Weighed quantities of phenol (14(1 Rra:..s) and ar 
senmus o.xide (80 grams) were heated fgether, and in all eases it was 
observed that the reaction began at Illll", the mixture boilitijj violently at 
that temperature. The tliermometer gradually rose to 111;) , "wh.'n a thick 
cloud formed in the Bask. The temperature remained constant at that 
po.nt for a lew minutes and then gradually rose to a .naxirauni, where it 
was kept for about five mim.tes: the heating was then stopped and the 
contents of the flask allowed to cool. The mixture of plieaol ester and 
arscnious oxide, the latter of which is ^ery solubl,. in the est.T 'was shaken 
with benzine, causing the precipitation of the arscnious oxi.le'ilissolved in 
the ester. The mixture was then filtere.l and the benzene solution of the 
ester fractional under diminished prssur. The benzene distilled at 20 the 
phenol at 69-, and the ester at :m'. The ester, purified by redistilla'tion, 
weifcheJ Klo grams, representing a yield of 00 per cent. I'limyl arscnite 
IS a deep yellow, viscous liquid with a sspecific gravity of 1..5!l- it freezes at 
■11' and boils :i03- under a pressure of 30 ram. It dissolves reidily in 
methyl alcohol, benzene, ethyl acetate, or chloroform, an.l decompose, on 
addition of water, but not so readily as the fatty arsenites. It was found 
on analysis to correspond with the formula (C H O) As, or (O U ) AsO 

u- m-, and f Tohjl .lr.,o,,(«._One hundred grams of each of th"^ 
cresols were heated with arseuious oxide for thirty minutes in the appar-,- 
tus previously described, the arscnious oxide being separated by benze,,',. 
The hqoid was then fractionated, and the esters, after having l„.en pii-^'io ' 
and analyzed as in the case of the phenyl arsenite, were found to corre-, „h1 
with the formulae (C,H^.\Ie)^As, or (C H Me) AsO 

NapMhyl Arsenite.-One hundred grains of'naphthol were heated with 
30 grams of arsen.ous oxide for thirty minutes and the arscnious oxide ^m 
the ester separated by means of benzene. As yet, however, the pure e,te'- 
has not been isolated. 

Benzyl Arscnite.-One hundred grams of benzyl alcohol were heated 
with 35 grams of arsenious oxide for thirty minutes. The maximum tem- 
perature was 240'. The clear liquid was d,H..anted and 73 c.c of benzcie 
added, the arsenious oxide collecte.i, and the filtrate fractionated under 30 

«lc-oh,.i, tlH.,.,sol,o, t,. drop in , 'ii., '""""{"•'•^ "> «"»" only .he 

tlnr .ha„ the side U,L of ,h, So" hic Th " f" '"""'' '" '^^ "'- 

«ppara,>,s, thcro bcm,- n.v r .„ffioh.n; 7, "^«" '''™<"' 'emainfd in this 
'he syphon :ma,..„„o„^::::„:"i!J™;L;""'~' °f """'^ P"«'-.« to cn„„. 

n;o™,^r:::;:;';..;:,ri:^-;r^::z,'r r- >''"-"^ '-^ - 

te™inrro„:;tri;^;f::rif::r"V'' ''°'"°'-— » d. 

>.-vornl ti„,e, in ord.r to re, novo „,1 'ral / • "°'™"^ "> ■""" ""^ 
".V -lubie in then,, in ^o^^:^ :^7^::^':^'l':^' w,.ieh i. .end. 
Sram, of the ester were decomposed by 3 e e „f » '^^ ""'■ '^''°"' " 
slum hydroxide (eontainin.- ™ 1 , "' " "'' '" ''■''■ •" POt«»- 

and the whole di ed /„ ee Tw" "',"""""' ''^'™^'''^ P" "'->- 
for analysis; to one sanUo standard r™"' "' " "•"• *•■"=' ""<■ '"k™ 
the for^itio'n of a Sae of ii •:;"', "", '''''!'' '" ''"'''' "■— "y 
Clear yeliow solution^the C^U,: -^tfi::; aTo,'"" rTT "' " 
tlien cooled, acidified nitli «,ilr,l,„r!„ . • i , ,., ' ""^ mixture was 

The excess of iodide n 00 1^^ "f r'' ' ""'' ■'"""' '" •'"" "' "■'■■ --^'o- 
sulphate, nsin, starlh'L iTd^ to ' 't;';! l!?;':;' ;«■""!'■'-"-" -io. 
multiplied by five represent, tl, . n ..."'"""'"J' "^ thiosulphate necessary, 
subtrlcted fL . r Sla a .n'^f ''d 'll"" """'• "^'''^ """""'^ 
eessary .„ change both fte ph t„ tri dth t ' fT '"^ ™™"' "^- 
to arsenic oxide To dcternHnetl '"' ^"P'^-ol and the arsenious oxide 
the ester, an excess ofrnZd ,'""''' "' """'"'' "'=''''■ P'-^'^t in 
«eond smple and this excess waftf'""" ?'"""'' ™ ''^^'"' '» ""■ 
Phate. The'ratiobeu:enthedihr„™t trj;::\='»"t'' ""™"' '"'' 

the differenb„Tt,e amount cZhd'"\'° """ '"'"'""' ■i^'ermination, 
the relative .uantlfi „f ^n ^^^ ^.^^ :" " f P;"-"'- ^™"' "■^- ^"t. 
;. the ester, were obtained, a^X't ^ ^^i;:^ :r:;e^rS: 


staoda'r7io"in:.' '"" '"""""■ ""^ '^^ '» ^^ ^^"'™>-' - ^^-^e.c o. 

droxii^i:: i!^zf^~ ^:-! -■ - *-^ - pota... h. 

Ten e.c. of this solution were found to be 'equivaient' 'to B-S cL" of 

standard iodine. 

(c) To 10 C.C. of solution (1), 10 M. of solution (2) wen added and 
found to require 32-30 c.c. of stanilard iodine. 

Thus 10 e.e. of solution (1) r,..,uire:| 27-7.1 c.e. „.' iodine, and 10 e.c. 

*'"''°? '^' '^"^"''"^ ''■^'' "■"■ "' '"'^''"''■■' in ■■'" :'--2--' c'-c. as conipared 
until 32-30 e.e. when miied, a dinen^nee which is well within the limits of 
experimental error. 

In order tu see if arsenimis „ji,le e;in be deteriuinel necurntely in the 
presence ot phenol by means of dichron.ate, experiment:: similar to -hoso 
made with iodine were carried out with it. Ten c.c. of soluticui (2) require,! 
BO, c.c. of dichromate, a mixture of 111 c.c. of (2) with 10 c.c. of (1) re 
quired 8.0!) e.c. of diehromate, or an error of leas than 0-3 per cent. 
Pro,,crtks of the Arscnitet prepared from Phenol anJ its Homolognea. 

Phenyl Ben.iyl oTolyl ni-Tolj-l p-Tolyl 
arsenite. arscnitc. arsenite. arsenite. ars"?iite 
Vield (per cent) 

(u) with Soxhiet 60 100 w 91 <« 

(6) without Solhlet .... nil nil nil „ii „|, 

Specific gravity \,r,Q 143 _ ,^- ^.^. 

Refractive index* l-57'» 

Boiling point * under 30 in.m. 30.'!' 2.5.5° _ 346" _ 

Freezing point — 31° --36° ' 

*''"'"""■ J'^''li>»- I'liiP <lark brown dark brown brown 

The blanks indicate that the refractive index is greater than 1-62098 
the limit of the prism used. ' 

t Where the boilin;; point is not given it is above ,360°. 

All these esters are soluble in methyl and ethyl alcohols, ether ben- 
zene, ethyl acetate, or chloroform, and are decomposed at once by water 

This method of esteriflcation is being carried out with arsenious oxide 
and the diliydric and trihydric plienols, but the quantitrilive results are not 
ready. A successful attempt has also been made to form similar 
esters with hydroxy-aeids, methyl salicylate being heated i.ith ar.enious 
.•xide. The products of the reaction, namely, water and an ,.il. were driven 
«p into the Soxhiet where the latter decomposed, liberating arsenious oxide 
The oil boils at about the same temperature as the methyl salicylate and has 
an almost unbearable odour. When the dehy '.rating agent is used in the 
Soxhiet it is expected the new ester will readily be separated. 

Experiments have also been tried with arsenious sulphide in place of 
the oxide and a small yield of an ester obtained, presumahlv of the com- 
position R.,AsS,, but the upper portions of the flask and the" condenser be- 
came coated with the oranj;e-eoIored arsenious sulphide, showing that de- 
eomposition had occurred, rhe work is being continued. 


LXXVII. — Sonw Esters of Antimony Trioxide. 
By John Fkancis MaoEet. 

In s previous paper, Lang, MacEey, and Gortner (Trans., 1908, 
93, 1364) described a method for the esterification of arsenious 
oxide with the fatty alcohols and with phenol and its homologues 
by removing the water produced by the reaction : 
6EOH + AsjO, = 2KsAs03 + SHjO 
as quickly as it was formed by means of anhydrous copper 
sulphate placed in a Soxhiet tube attached to a flask containing 
weighed quantities of the reacting substances. By this means a 
large number of esters, which other methods had failed to produce, 
were prepared in quantity, and found to correspond with the 
general formula RjAsO,. In the present work, attempts were 
made to form the corresponding esters of antimony by five different 
methods, namely: (1) heating a mixture of the alcohol and 
antimony trioxide in a flask to which a reflux condenser w« 



att«ch.d; (2) heating a mixture of the alcohol and BotimooT in a sealed tube at 10(P; ^3) .baking a mixture of the 
alcohol and antimony trioxide with anhydrous copper sulphate in 
the cold; (4) heating a mixture of the alcohol and antimony 
trioxide in a flask with a Soxhlet tube, niled with anhydrous 
copper ...Iphate, attached; (5) heating a mixture of the alcohol 
antimony trioxide, and calcium tu-T,ings in . flask with reflux 
condenser. A comparison of the y.elds obtained is giv a in the 
sequel. By means of these methods, and more particularly the 
last two, esters were formed with methyl, .,thyl, propyl, wobutyl 
amyl, and isoamyl alcohols; also with phenol, „, m-, and p-cresols 
The esters were found to correspond with the general formula 
RjSbO,. In all cases the yields obtained with antimony trioxide 
were very much lower than the corresponding arsenites previously 
deccribed. ' 


(i) Heating with reflux eondemer oniy.— Weighed quantities of 
methyl, ethyl, propyl, Mobutyl, amyl, and ijoamyl alcoiiols were 
mixed with an excess of antimony trioxide, and heated for different 
periods of time ••.. a fla..k to which a reflux condenser »as attaclie-i 
After five hours' heating not a trace of ester could be detecUd 
in any of llie leactiug nii.xturea; but, after fifteen hours' heating 
evidence was obtained that, in the case of wobutyl and uutmyl 
alcohols, esters were formed, but not in quantities sufficient to 
permit of iheir being isolated. That an ester was formed was 
shown by pouring the clear liquid into water-free beniene* in 
which antimony trioxide is very sparingly soluble, and filtering 
By this means any antimony trioxide merely dissolved in the 
alcohol was separated, and, if antimony were found in the tbove 
filtrate, it must have been present in the form of an ester The 
filtrate gave a distinct precipitate of antimony sulphide when 
acidified, treated with hydrogen sulphide, and warmed; thus 
whilst there is no doubt that both i.obutyl and Moamyl antimonites 
are formed under these conditions, Ihe yields are so small that it 
IS impossible to isolate them by fractional distillation. 

(ii) Heating in sealed (u6«i.— Quantities of the various alcohols 
were mixed with antimony trioxide and heated for six hours in 
sealed tubes at 150°, but in no case could the presence of any 
c -er be detected. ' 

(iii) Using a dehydrating agent in the eoW.— -Weighed quan- 
* Gr..t difficulty wu found in obtaining beuMn. free from w.ter 
The rwnple. of benzene at hand were found to contain enough water to decompose 
partly the e.t.r. form.d. It was necmry to heat the bennue with calcium 
tuminga for about forty-eight hour.. 

maikey: some esters or antimony tkioxide. 606 

titin uf the varioua deohols were mixed with antimonjr triozid* 
and a considorable amount of anhydrous copper sulphate. The 
mixture was then placed in a sliakiug ninchinc for three days at 
room temperature. In the case of ijubutyl an" tuoamyl alcohols, 
esters were farmed, but not in quantities sufficient to enable them 
to be isolated. 

(iv) llralinij wilk Hiirlilrl (Utachvirnl and anhi/rlror.s cupper 
mlphate, — Thirty (-rama of antimony trioxide wore added to 160 
grama of ijoaroyl alcohol iu a 260 c.c. flask. An ordinary Soxhict 
tube vas connected with the flask, and a condenser, fitted with 
a caUiuui chloride tube, attached to the Suxhiet, which contained 
a lar>;c filter paper filled with anhydmuti copper sulphate; the 
mixlii u in the llaak was heated to boiling, anil the water formed 
by the reaction was absorbed by the copper sulphate, which turned 
blue as soon as the first drops of condensed liquid fell on it. 
Heating was continued for about fifteen hours i after cooling, the 
clear product was poured off from the antimony trioxide and 
fractionated under diminished pressure. About 108 grams of 
the ester wrre formed, representing a yield of 13 62 per cent. 

By this method enters were formed with methyl, ethyl, propyl, 
I'subutyl, amyl, and tnuaniyl alcohols, as shown in the following 

Table, of yields by varioui method*. 

Alihydmii.s copper -sulpliatf. 

Kt>flii.\ III Kfiilcd .. ' s 

Esiler. comlfiisci-. tiitius. In Hm cold. In Soililot. 

Molhyl nil iiii iiif 8 09 per rani. 

Kthyl „ ,, ,, 3'00' „ 

l'ro|iyl ,. „ ,, 13-3 

iwBulyl Lace „ triwo l.^i-ll 

Amy] iiif ,, nil 13-16 „ 

isokmyl trace „ trace 13'62 ,, 

" It will lie noticed that the siime e.^lrentely low yield ol ester with ethyl a'cohol 
occurred iu th ; preparation of the csterM of arsenioiw acid, described in a fornicr 
Iia[)er, where a {loaaible explanation wail advanced. 

These esters were purified and analysed, the following methods 
of analysis being employed. 

Method of analysis for esters uf the aliphatic alcohols (usiny 
isoatnyl antimonite as an t,ratnplt). — Two methods were employed 
for the determination of the antimony. In the first method, 
iodine was used to oxidise the trioxide to the pentoxide, but was 
not found to give accurate results, so that the following procedure 
was adopted. 

r0057 Grams of the eater were decomposed by 5 c.c. of con- 
centrated hydrochloric acid, SO c.c. of tartaric acid solution vera 


•dded, >ad the intimon; wu preeipitet«d from thia Klution by 
muoi of hydrogen nulphido; the •ntimony lulphide thut formed 
WM converted into Mtimony totroxide by treatmoat with fuminc 
nitric acid and ignited: 

10057 gave 040^4 Sb.O,. Sb = 31-87. 

l-'isHaOjSb rwiuires 8b- 31-48 per cent. 

Thii method of analyiia gave the foUowug roiults : 

rcri:rliU(!r of Sb 
_ I orccnUKe „l inlculiilc.l frnni tlic 

„ . ,"• Shloiiml. lunimU B.HhO,. 

^("::;::::::::::: r' ??j 

Expressing these u gaits of autimoiuus acid, we arrive at the 
general formula BaSbO, for them. 

Propertit, of the E,Ur. of the Alip/mtic Alcohol, with Antimony 

1-025'*^' a»<»mom<<! is a colourless liquid, b. p. 65°, ip. jr. 

Ethyl antimonite is a colounesa liquid, b. p. 116—1200 

i'ropyl antimonite is ft yellow liquid, sp. gr. 1042^ b p 
143°/ 30 mm.; at atmospheric pressure it decomposes into 'propyl 
alcohol and antimony trioxide at 200°. 

uoHutyl antimonite is a yellow, mobile liquid, sp. gr. 1058, 
b. p. H4°/30 mm.; it decomposes into isobutyl alcohol and 
antimony trioxide at 250° under atmospheric pressure. 

Amyl anttmonile is a yellow liquid, SD. er 1079 I. n 
no°/30mm. ' r s - "w, o. p. 

isoAmyl anlimonite is a yellow, mobile liquid, sp. gr 1081 
b. p. 163°/30 mm.; at 250°, under atmospheric pressure, it 
decomposes into antimony trioxide and .soamyl alcohol 

All these esters are readily soluble in absolute alcohol, ether 
chloroform, or benzene, but decompose immediately on addition of 
water into antimony trioxide and the alcohol from which they were 
prepared. ^ 

The fifth method was not employed for the aUphr.tic alcohols. 
Esters obtained with Phenol and its Homohyttei. 

For the cstei-ilication of these substances with antimony trioxide 
two method were used, namely, heating the mixture of phenol 
and antimony trioxide in a flask to which a Soihlet apparatus 



conlkining anhydroui copper sulphite wu attkched, and htatiDi; 
Ihe mixture o{ phenol and antimony trioxide with calcium 
turnings in u flask under a icflux condenser. By those means 
esters were obtained with phenol, it-, m-, and /^L■resul8. In the 
case of phenol and /H-crcsol, uiiHUcccssful attcmi>*<' to oMain the 
esters were made by heating these mateiials with antimony 
trinxidc with a reflux condenser only. 

rhrtiyi ti/ilitnunite. — One hundred and twcnty-tivc grams of 
phenol and '25 grams of antimony trioxide were heated in a flask 
titted with a SoxhlO apparatus containing anhydrous copper 
sulphatj. Clouds of sicam began to form at IUO°, indicating that 
the rea.'tion began at that temperature; the thermometer rose very 
quii'kly to the boiling point of phenol, and remained very close 
U> that temperature (varying from 170° to 18U°) for about seven 
hours, when the temperature gradually rose, ^nu, after liftccn 
hours' heating, reached a maximum of 290°. This maximum is 
the highest temperature to which the ester can be heated without 
decomposition. The mixture of phenol, ester, and antimony 
trioxide was then jhaken with water-free bei.zeuc, causing the 
precipitation of the antimony trioxide dissolved in the ester and 
in the phenol, and, after liltering, fractionated under diminished 
pressure. The benzene distilled at 30 . phenol at 70°, and the 
ester at :ioO°.* The ester was purified by dissolving in benzene, 
filtering from any antimony trioxiue, and separating by passing a 
current of hot, dry air over the benzene solution placed in a 
vacuum desiccator. The crystals thus obtained had no very 
detinite melting point owing to the presence of some phenol. To 
remove the latter, the crystals were heated to *JO0° for twenty 
minutes, dissolved in benzene, liltered, and crystallised as before. 
The crystals thus obtained were light brown iu colour, and melted 
at 13°. At 18°, jihtnyt antimonite has a specific gravity of r621, 
and boils at 2.10°/ 30 mm. It dissolves readily in absolute alcohol, 
ether, chloroform, or benzene, and decomposes on addition of water 
into phenol and antimony trioxide. On analysis, it was found to 
correspond with the formula {CeH5)jSb03 or (CeHjOjjSb. This 
ester was also prepared by heating a mixture of phenol, antimony 
trioxide, and calcium turnings in a flask fitted with a reflux 
condenser. By this method, a yield of 62 per cent, was obtained 
after two hours heating. By the former method the yie'-' \ not 
more than 40 per cent, after fifteen hours' heating. 

0-, m-, and \t-Tiiiyl A nlimonilts. — One hundred and i „raiiia 
of each of the crcsols were heated with about 30 grams of antimony 
trioxide, using methods described above. The excess of 

* £T«n St 250° th* wter sutfers partial decumpotition. 



.ntim,my t,i„xi,lc w« «,p»r.U.d by m«„, of b,n«,n., (Ilt.r«l. ..d 
l,r o,u.r. fra,.t.o„«lod. After purific.tion they .naly^d! .nd 

((.".•trr,0),Sb. o.l.,lyl onlimnmu cry,t.lli«. in d.rk bjown 
C.V ; „/ .!;t* '"'";""«'" '■ • yollow liquid, with . .p«ifie 

13 .dark 1, .,„n, crystalline «,lid, molts .t 14=, |,„ . .peoific 
gravity „f l,i.j,5. and l.oil, ,t 345-^/30 mm ^ 

WlulHl in the ra.u of the aliphatic alcohol. . drying aRcnt in 

and allow the ,I„,|, only to drop ha.-k into the mi.turc, no 
y.»K «K..n «.v neces,,ry with the phe .» practically no 

...mcLT*^"! ^"""';' '■'""'''""'' '" "■" '*""'■''"• "'Of^ hoiog -v.r 
In , ""\"' r"'" '•""'''"'" ♦" '""'•'> "■' -yphon attach 
nent t-, come into play. The u,c of metallic calcium, it wi" be 

r,/' T, °.'''^'"- ■""''' "' ■»'" "■" '''o Soxhiet method 
Krama of phenyl antimonitc, 6 c.c. o'. pota-ium hydroxide fco„ 
ta,n,ng 7 gram, per ,0 c.c.) were added, and iulJZJ^lt 
tarUnc and. The ,...^c.» of t.rUric aWd wa. ..eutr.lii^ b! 
.^d.umc.r„onate and .^ ,, „j ,„,;„„ ,y, earbonat^^ded' 

The whole was diluted n ,500 c.c. and two .ample, of 20 ec 
each were aken for an.Iysi.. To each. 20 c.c. .Card Tod^e 
was .-..Idcd m exce., (shown by the formation of a prec piti^ o 
r-,odophcnol and the ap,,earance of a clear yellow .olution^ Z 
temperature being kept a» 65= The mixture wa. then eLu 

00 ec Tie ^ TT' '"'' "" ''""'«° with wa^r t^^ 
auo c.c Tl.e tn-iodophor ,1 was filtered off, ind the excess of 

th o ulphate, us,ng starch a» indicator. The quantity orthio 

oline ^h"r""'' '""'"'""•' ''^ '''' "P-"'"' t'"= qu«ntity of 
iodine that was in excess This quantity, subtracted fro, tZ 
original amount of the iodine added, gave the amount nZj^l 

o change both the phenol to tri-iodophenol and Z ^^Zl 
tnoxide to antinony pentoxide, antimnny 

Owing to the presence of tartaric acid in the above, it was 
impossible to estimate at all accurately the antimony tri^ide by 
means of dichromatc. The method previously described under 
x,,.n,y! antmiuiute namely, weighing the antimony as antimony 

tT ' ""' "'",'■ """^ P^^'^^'^S'^ "f .ntimony, „ det,rm"ned 
in Uu. way, wa. found to be 3010. The .tr.n|^L of the^^n. 



heinK known, thr numhrr of c.r. of inHine MjuiTalrnt 'o 30'tO prr 
cent, nl antimony wat militrarted from the iodine nhtainml in llio 
previoua dct«rminstinn. tlio difference being the amount romliinrd 
with phenol. 

To teet the accuracy of thia method, eetimationii were made 
with weighed quantiliei of (a) phenol, (6) antimony trioxide, (c) 
a mixture of thee*. 

(ii) To 08826 gram of phenol, 5 c.e. of a eolation ' potanium 
liydroxide (7 grami per 10 c.c.) were added, and the whole diluted 
with water to 280 c.c (1) 

10 c.c. of thii solution were equivalent to 28'90 c.c. of iodine. 

(h) To O'SOOO gram of antimony trioxide were added 25 c.c. of 
a aaturated solution of tartaric acid ; aufficient aodium carbonate 
to neutraliae the exceai of the tartaric acid and SO c.c. of aodium 
hydrogen carbonate were then added, and the whole diluted with 
water to 600 c.c (2) 

100 c.c. of thIa aolution were equal to 13'70 c.c. of iodine. 

(e) Tv iO c.c. of (1) were added 100 c.c. of (2), and together 
required 42'4S c.c. of iodine. 

'"'lua 100 c.c. of (2) required I3'70 c.c. of iodine, and 100 c.c. 
1) required 28-90 c.p. of iodine; in all, 42'60 c.c, aa compared 

,li 4i45 c.c. when mi.'ced, a difference of leea than one third of 
1 tr cent., which ia well within the limita of experimental error. 

Propertiei of the Eilert of Phenol and ilt Homologuei. 

Phdnyl. o-Tc.lyl. ,/i-Tolyl. fi-Tolyl. 

P.rcentto-e yWd 40-0 48-2 48-3 408 

B. P./.30 mm 2B0- .•ir,2' 300' 34.V 

Sp. L-r 1621 1-480 l-47.'i 1-495 

ii.v- "° '«■ - '*" 

The preparation of other e-.ter« from the oxides of arsenic and 
antimoi^y anil rompniinils rontaiiiing ali-oholic liydroxyl ia being 
proceeded with in this laboratory. Attempts are also being made 
to overronie the difficulties experienred in obtaining the eatera of 
sulphoarseniotis acid in sufliricnt quantities t.o allow of their com- 
positions being determined. 

I wish to express my thanks to Professor W. K. I.ang for 
suggesting this research, and for the interest he has tikcu in the 

Chihical Laboratory, 

Univbrsitv of Toronto.